S-Prihadi

Guys,

If anyone can share its location, many would be grateful.

This ground is critical for your engine computer.
However, too bad for W212 with M276.9 3.5 liter Normally Aspirated or M276.8 3.0 liter Twin Turbo and M278 or M157, MB WIS has no actual location data.

Typical M276 3.5 NA engine



I tried on Star Finder, no luck too.




WIS will lead you to nowhere for W16/5, no LOCATION information.





What modules use W16/5 ground points ?

01. Engine computer direct from W16/5
02. Engine computer , from X26 connector which source it from W16/5 too.
03. DataLink Connector, or OBD port as we call it
04. Aircond compressor if car version is a hybrid W212.095
05. Hybrid fuse and relay box K40/10
06. A module called N129/1 power electronics control unit. I believe this is for hybrid car only.



On newer W213, Star Finder only give general location for W16/5, at least something



On W211, W16/5 is not in use.



On W205, the W16/5 general location in given and I am sure it can be found as this 3rd party information is a photo with decent resolution.




Since I am on a right hand drive car and 3.0 TT M276.820 , it is a more nightmare for me to find W16/5 as the WIS is more left hand drive oriented database


My guesstimate for W16/5 location for a Left Hand drive car with M276 which has its computer on the LEFT side of engine or in USA speak : "at driver side" .................,
I would think the location of W16/5 would be at the car frame. Somewhere W70 and W9 is at, maybe at the opposite side of that metal frame or engine side and not wheel arc side.


Why I guess the location is there is because :
AA. Left hand drive car Front SAM is at the Left side of the car.
BB. M276.9 3.5NA engine computer and the X26 connector is also at the left side of the engine.
CC. X11/4 datalink connector or OBD port is also at the Left side of the car cabin aka USA side driver.

So it would have to be at the left side of the engine bay vicinity if not at interior ?....as to be as close as possible to its "consumer" and even WIS called it as :
LEFT MAJOR ASSEMBLY COMPARTMENTS ELECTRONICS GROUND POINT.

Original documents are in German and translated into English for us, so I would not really believe what is written as 100% accurate compared to a US or UK native English speaking/writing engineers.


Any help is most welcome and thanking you in advance.

Left Coast Geek

lets see what I find in my copy of WIS-ASRA (a bootleg VM from fleabay)... Il'll use our 212.288 E350 4matic facelift wagon as a base car.



ok, the document that claims to show where W16/5 is seems to be missing from my WIS, its PE00.19-P-2800DAE and its greyed out in each index I find it referenced in, and not present.

S-Prihadi

Yes, the same GREYED out if you access it and still no location photo........with my 2018 and 2020 WIS..........

CaliBenzDriver

Near this location...


Painted GND Post nearly impossible to clean?

It's located on the firewall towards the center near the brake booster.


ECU is spaning the Engine-GND & Chassis-GND :
-- Some of the ECU gets alternator GND directly from each valve covers front.
-- The painted GND Posts are all located on the Chassis-GND.

The car Chassis is normally grounded to the Engine block by the single strap at tranny... when that whole-car connection develops a drop-voltage under load, the Chassis and Engine are then using different voltage reference... ouch!


How to delete some of MB chaos :
Do not span unreliable GND for reference
Fix the chassis GND to be reliable
Poor connections need to be eliminated painted/solderless
​​​​​
GND must be trusted to be at identical potential or else the poor reference seeds chaos. GND is both extremely important and extremely simple.

​​​​​

S-Prihadi

That make sense Cali.
If true MB stay accurate to the definition of the LEFT MAJOR ASSEMBLY COMPARTMENTS ELECTRONICS GROUND POINT, it has to be at the LEFT side.
If indeed the W16/5 near the brake booster, that is the mirror copy of the yet-unknown ground cable a tiny bit above my battery tray of the LEFT side of the car,
since I am a RHD car and my brake booster is at the RIGHT side of engine bay and my battery is the one on the LEFT.

Ignore W16/6 label below, W212 gasoline does not use W16/6 as far as I seen on WIS.

CaliBenzDriver

So it looks like we have this ECU GND located on the same spot regardless of driver handedness (LHD / RHD).

Lucky you can clean yours by only removing battery unlike most LHD cars.


TEST : GND POST MEASURES
It would be interesting to measure the condition of that Painted GND Post - Either or both the static resistance and dynamic drop-voltage in relation to the engine block.

The block is our standard reference because ALT normally supplies 100% of the load, all returning to the block.

S-Prihadi

So, in theory, 6 and 8 cylinders gasoline engine W16/5 would be total 5 wires. I initially thought only 4, because I never counted the OBD connector yet.









1st to 3rd wires are 1.5mm brown-white direct to ECM , this is main ground INPUT for ECM. So all 6 injectors which is some sort of special circuits, being the only 200 volts section of ECM get its "ground" from here.
4th wire 2.5mm BROWN - shared main grounds for all COPs
5th wire 1 mm brown-white for OBD /Datalink



===============

Now, specific to COP . It is the only component with 2 grounds. 1 from W16/5 and 1 more from W11.
The last time I current clamp W11 for the M271 engine, W11 does not carry current at all, hence I call it noise control ground and it is obvious W11 at COP it is smaller 1mm BROWN wire compared
to 1.5mm of brown-white for each COP which comes from W16/5.



Note : M271 is a dumb COP where its driver is at ECM and not at COP. So only 3 wires into COP.
1 is W11 and 1 more is MAIN negative feed from ECM which is equal to W16/5 of our M276/M278/M157 4 wires COP.

If one wrongly assume that W11 is the MAIN negative ground for ECM, because it is conveniently located to access, he will get into trouble.
This also shows that MB is not using engine block which is where W11 is located, as MAIN negative power source.
MB still uses the better MAIN negative/ground, which is chassis-Ground, the W16/5.
Do you agree Cali ?

S-Prihadi

Aha, now that you mention the ALTERNATOR thingy......................
The only way to verify with certainty for W11 and W16/5, I need to current clamp both W11 and W16/5 wire sets and see if W11 on our engines is a indeed noise control ground only or current carrying conductor ?
I have to do visual inspection of the assumed W16/5 near my battery tray to do its wire counts, size verification and colors verification.

CaliBenzDriver

Thank you Surya for extracting rock-solid material and the interesting information from it. Now we know how this bundle is populated without any speculations.

Yes I do, I agree this "Painted GND Post" on firewall is MB using the Chassis GND for this ultra noisy circuit: Spark coils and Piezo injectors and ECU power.

However I would not call the "Chassis GND Post" a "better negative GND" - It is not.... the best GND is the shortest back to what supplys power to the load. The power that leaves must come back: it's an electron loop!

This circuit is a seriously poor choice that helps make things worse than could be, for FREE/peanuts.
Such that...:

-- The COP/Injectors/ECU GND path is made needlessly extra long through the amazing chassis GND strap to ALT. The equivalent length add a resistor in theses circuits.

-- Everything connected on the Chassis side of the main GND Strap is blasted by nothing less than every spike created by COP/Piezo... OMG.


> QUICK FIX :
The best impedance/resistance is shortest path, the most non-glitchy way would be to have (Cop/Piezo/ECU) short GND path to the block, not chassis side of main strap.

The quick fix is to strap that one particular GND Post directly back to the block. In essence that additional strap is going to be the missing secondary backup path away from road salt oxidation.
This should provide better power with less drop to run the engine coils spikes and reduced noise forced onto all the Chassis loads.


> Ridiculous POOR CHOICE :
The way this noisy load GND circuit is wired helps introduce noise into unstable/soft-crashed F-SAM and create weak spark/injection circuits...

We can experiment flipping this snowball from bad to reliable.
I wonder how to be able to remove the Painted nut on LHD US cars? (I am not dumping the Brake Booster, Master cylinder...)


> GND TEST FUN :
You said the DLC/Diag. GND through that mystery Painted-post, true?

Then scope the DLC GND with reference to block, not chassis. This should show the loss and noise in main GND strap.


++++++ You guess WHY?
Why did the best engineers go through the trouble of extending the GND of engine Ignition + Injection - They could have perfectly grounded right there at the block without extending out to the chassis... Why not?

++++++ GND: Signal vs. Power

-- Signal GND is one you trust to be a reliable reference for a signal refenced to GND, such as a slow LIN signal (Not CAN that is floating between two lines).

-- A power GND is a heavy line that can pass current and which cleanliness is not necessary. Example : the 100A-Fan can eat tall spikes But not so much for a SAM that is tasked with measuring voltages.

> Power Surge failures:
Imagine where the ignition/injection spikes go when the main strap is aged and the Painted-Post is becoming gradually more marginal ?

S-Prihadi

Lets call the 2 types are grounds in this DC as :
AA. Ground for power (-12V) or Neutral so to speak in A.C system.
BB Ground for noise or spike protection, or earthing for to speak in A.C system.

The engineers were after equal voltage drop when engine is running, hence the Ground for power (-12V) uses car metal body.
From battery negative it is the W10 ( 1.5 inch brass stud at strut mount ) as first MAIN ground for power.
From the same car metal body, the engine block at starter receive -12V via that W-TF wire under the car.
The alternator uses engine block as conductor for its -12V to supply local engine demand and send excess to battery , again via W-TF.

That is why W11 is not supposed to be Ground for power (-12V) as far as ECM is concerned, but is a Ground for noise or spike protection.

This video explains quite well :


The W16/5 male stud is not painted at the important contact points, only the top hat of the female nut is painted. The contacts of eyelet top and bottom are good.





========================

The total wires QTY at my W16/5 wire set.




=================



Here are the measurements : I do not need to do load test, it is confirmed a 100% that W11 is a Ground for noise or spike protection only and not the actual working/loaded conductor.












1.068 Kilo ohms

















So, we all must really watch out and super maintain for W16/5, as that is the bloodline of the ECM.
Damn, I feel so bad for LEFT HAND DRIVE cars, the location of the W16/5 is so difficult to access and in actuality, W11 the assumed important ground is not a power delivery conductor/wire.



ADD:
I love using strings ............

S-Prihadi

WTF is MB trying to do with my engine's COP wiring ???

I verified true wire size on the connector, schematic is not a typo. This was why I been all along in the past thinking that W11 is the Ground for power (-12V) and
W16/5 Ground for noise or spike protection, simply from chosen wire size by MB engineers ........... until I really pay attention today and compared it with M276 3.5NA .








Why use 0.5mm tiny size when it is the real Ground for power (-12V) ?
And then the W11 which is supposed for noise control is getting 1.5mm, tripple size ?? DUGGHHHH.......


Now COP #4 has a 0.2 to 0.6 ohms extra resistance, without a LOAD test. Simply continuity.
0.5mm is CAN-BUS data cable caliber, it is not for any sort of power delivery and to a COP with demand of 8.5 AMPS !!!

By a simple shake-shake & push-pull test on that damn super small 0.5mm wire, the problem is with the crimping and not the splice upstream.



I have to replace all 6 of these damn small wires to 1.5mm ASAP.
And this connector is not easy to remove its female pin.... what a sad affair !!!

Westlotorn

Once again, very thorough advice Mr Prihadi, Maybe the small wire is because the electric pulse is so short time wise that a small wire can handle it fine? If I read correctly you think the W16.5 is on the driver side of my American spec car, that would put it under the Master Cylinder and maybe under the Fuse block. I have traced brown wires in the loom that go through the fire wall headed that way but have yet to find where they bolt to the body. On the Pass side of my car I did find a ground to the body not far fromt he alternator and cleaned it up well with no change. Been tied up with GrandChildren the past couple days, no car time. Hope to get some time today. I did pull the harness apart again searching for issues.

S-Prihadi

Man, me old eyes suffered un-pinning this connector. I hate SLK 2.8 locking ears.

The crimp is okey, it is the inner biting tounge of the female terminal of COP4 pin #1 which is not too bite-y anymore. I have to admit, it has been disconnected exceeding total 30 times.


I recalled a member here did mention some sort of "latch" when he did the COP connector. Here I am explaining as best as I can.
I have spare COP connector, so I can practice first.





Trying to flatten the LOCKING EARS is no fun. I do not have original tool for it, but had the cheapo one and none is a good one for SLK 2.8 locking ears, two of them, one at top and one more at botton.



The 2 locking ears




COP4, 5 and 6.









I will work on it tomorrow. Today me eyes so tired using magnifying glass too long working on this Dumb-Azz connector.

Westlotorn

After taking apart the connector for the Fuel Injector at #1, I would not do that again. Take one strand of wire from a stranded copper wire. My one strand measured .005 inch when I checked with my Mic. Take a piece of flat steel and a hammer, hit your piece of stranded wire, tap it a few times, you can watch it get wider as you tap. Measure as you do this. I make my piece .001 thick. Slip this piece into the female side of the connector in the pin you want tighter. If the strand is wide enough it will lock in place as you install it. Snip it off flush with the top of the female connector. Waa Laa, female connector is now tight with a good clean copper shim in there. The idea is to have a good connection. Many ways to get there. Biggest danger doing it this way is the piece falls out and shorts something so make it wide enough to slip in tight and stay put. The test fit pin works well to push the copper shim all the way in. I actually tapped one of my test pins to make it thinner to use for this install purpose. Should this be in the book of solutions? Probably not, but do you want it to work? It can make it work. I say this because on the injector connector after removing the wires from the connector and pinching them tighter I fought for over an hour getting them back in.

Edit: A couple weeks after posting this I make more Shims if you will to tighten the connectors. Found if you take a thicker strand of wire, I now use a .025 thick wire and pound it flat for a measured thickness of .008 or close to that it works well in the pins with wide pins. Also, to keep them in place after install I started bending a J shape into the end that inserts. Place your test probe in the J and use the test probe to push your shim J shaped wire into the female connector. As you remove your test pin the J shape now holds the shim in place and your connector is now tight. Taking the connectors apart just took too much time and I am not sure it actually worked to tighten the pin up. The shim seems to work well and it is easy to do.

CaliBenzDriver

Ok, I think we've got good hard evidence of a poor GND.
Check this out :


Can you spot anything wrong? 🤘


obstruction by multilayer pins


PINS prevent basic contact on single eyelet posts !!


oxidized discoloration


painted lip = NO CONTACT!

> Dealing with this GND snowball:
I believe this special "Mother of all GND" is the elusive gremlins factory we have been looking for a long time.

This VIP GND is responsible for :

• poor piezo fuel circuit
• poor COP spark circuit
• poor ECU circuit

The eyelet poor contact surfaces and joined circuit sends coil spikes into the ECU ...


What are the issues layered here:

1. Shared noisy coil POWER and ECU SIGNAL reference
2. GND on wrong side of chassis-block
3. poor contact from eyelet pins
4. no contact to painted nut lip
5. poor single chassis GND strap
6. Coil/piezo GND circuit too long
7. harness mixed with noisy circuits

How to fix this snowball:

• GND with clean flat contact surfaces
• GND on engine side, not chassis
• GND ECU separately from noisy coil bundle
• (GND the chassis to block reliably, elsewhere?)

When your looking for cop/injector wiring issue, a short GND circuit is a prerequisite.


What is MB accomplishment :
Field test the robustness of advanced modules in extreme environments.


Furthermore, we need to rework all GND Posts with single eyelet: trunk AUX Batt GND run has single eyelet!
How: flatten or brake the eyelet pins such that both sides can contact GND properly.


> Borderline cheapo bush-fix :
In a pinch when you need it now, simply flip the eyelet over so the pins (moderetly acting as crush spring washer) face the painted nut side. The smooth eyelet surface is facing the post.
The dirty-fix outcome here is a vastly improved GND contact.


HOW can get our hands on this ultimate Painted GND Post easily ??
Can we lift out the fake-firewall for access ??
Can we reach through the top plastic cawl??

S-Prihadi

Enjoy your time with ur grandchildren first and foremost

S-Prihadi

Ok Cali, I will clean W16/5 again..... , now I have better brush kit

Westlotorn

If I can just find the dang thing I will clean it and trace all splices to it. I have measured .120mv voltage drop in this wire harness this morning but still do not know where it is attached to the body. I will pull the Front Sam out and hope to see it under the front Sam. I do not want pull the brake booster out to look
Still can't believe Mercedes would put such an important ground where it can't be serviced so I keep looking.

juanmor40

Ok @CaliBenzDriver , thank you very much for the photo. Even with it, I had a hard time seeing it at first glance. My initial guess to access it is to disassemble the wiper system, cowl, and attack it from the top. Nearly certain it is a similar step to remove the AC expansion valve. Not a chance the brake booster has to come out.

From @konigstiger , there is some documentation in this thread https://mbworld.org/forums/e-class-w...ml#post8728877

juanmor40

@Westlotorn , have you seen the ground point Cali posted? If not, you can shine a light from the top of the brake booster (brake fluid reservoir) down against the wall and it is there clearly smiling at us

Westlotorn

Dang, I should have checked in when I was working today I could have seen your tips. I was searching for a few hours this afternoon. Found a couple hidden grounds, of course they are in the photos provided this afternoon. I saw the grounds on the driver side attached to the frame rail. No access from under the hood so I took out the driver side wheel inner fender liner and gained access to two bolted body grounds. Now I know they are W70 and W9. I was hoping they were maybe W16/5. Oh well, they are now cleaned and ready for use. Tomorrow I should be able to get to the W16/5 by the brake booster. Man, I have been all in that area with a good flash light and not spotted any grounds or wires other than those going into the front fuse box. I am sure it is there, I will go back after it. With a 5 amp load I am seeing about a 70mv drop in voltage at my #6 pin in the smaller computer connection and about the same at the #3 pin in the larger connection at the computer. I need to set up my test so I can have free hands to wiggle the wires as I test to hopefully find the issue. It is in there, got to find it. I also am finding I need to test each test wire because they each have voltage drop that needs to be known and recorded for all tests to get good results. Thanks again.
Mark

CaliBenzDriver

Yes make it good with easy access behind AGM. Lucky you 👏


discolored contact tips of pins 🙄

At least, pound these pins flat or at 45° so they can act as spring washers against vibrations (Loctite would insulate the nut threads!)

CaliBenzDriver

I wanted to understand the COP circuit used in this engine

COP Pinout

The ignition coil has two GND...
one of the GND is Power return
the other is the Signal reference

We could expect "GND Reference" not to be super noisy unlike the power GND feeding the primary coil windings.

vs. this schematic version:

COP GND wiring sizes inverted
so now I am a bit confused with COP GND pin assignment

Now I want to understand the piezo injector pinout getting GND W16/5....

Ultimately I want to devide up noisy wires out of the W16/5 GND eyelet.
✌️

Westlotorn

I just typed this return and it seems to have vanished???? Dang computers or their operators can be a real pain at times.
I traced grounds again and tested again tonight using my 3 light set up which gives me a 5 amp load
No luck finding the X16/5 yet and I looked with a really good light.
Ground tests looking for voltage drop. I used the exact same wires for each test to avoid variance.
117mv voltage drop at the body ground on left side/driver side fender by ABS system accessed through the fender skirt removal
103mv drop at the battery post on the fender top/ copper ground stud for jumping the battery
103mv @ both valve cover ground locations
103mv @ the ground stud behind the pass side headlight on lower fender
150mv @ #3 small Pin computer connector in the large connector
153 MV @ #6 large pin in the small computer connector
135mv @ X26 Brown wire in the gray female connector
Largest Voltage drop is 50MV more than the best ground. Not sure 50mv loss under a 5 amp load is anything to worry about???

I tried to wiggle and squeeze the harness to make the connection go away, hard to do as every movement affects the pin connectors and aligator clips. Can't say I identified any iffy connections.

Thanks
Mark

.

S-Prihadi

Those are good numbers for 5 amps load.
I am now doing the same while I have access... I do 7.1 amps load.

Under maximum load usually we do not want to see more 0.2 volts drop.


Since your W16/5 is still attached somewhere, you can test its voltage drop and you actually did :
153 MV @ #6 large pin in the small computer connector << This is from W16/5 and is already a full wire length test if for ECM. But you must also test pin #2 and pin #4, they are also from (correction ): W16/5 just like the pin #6 you tested.